GRANT=Z01AR27015-04 Background. Many important cellular functions are performed by large complexes whose constituents function in coordination as working parts of macromolecular machines. Macromolecular complexes also play primarily structural roles as biomaterials in many tissues, including skin and muscle tissue. The goal of this project is to elucidate the structures, assembly properties, and interactions of complexes of both kinds, with emphasis on their functional connotations. (1) Energy-dependent Proteases. In bacteria and eukaryotes alike, most intracellular proteolysis is carried out by energy-dependent proteases. These enzymes generically consist of a proteolytic component and an ATP-hydrolyzing component with chaperone-like properties. We have focused on the ClpAP enzyme of E. coli, an attractive model system. In earlier work, we found that the protease, ClpP, consists of two apposed heptameric rings of 21-kDa subunits and the ATPase, ClpA, consists of a single hexameric ring of 84-kDa subunits. ClpA stacks axially on one or both faces of ClpP to form active complexes. These observations made a major contribution to formulation of the current paradigm, according to which the ATPase recognizes substrates, unfolds them, and feeds them into a digestion chamber sequestered inside the protease. During FY99, we focused on the large conformational changes that we found to accompany nucleotide exchange in ClpA ? either alone or as complexed to ClpP. The non-hydrolyzable ATPgS suffices to stabilize ClpAP complexes but not to support protein degradation. For this latter activity, ATP is required. When ATPgS is substituted by ATP, ClpA undergoes a large conformational change. Our working hypothesis is that this transition is mechanistically coupled to the unfolding of substrates during the active cycle and their translocation into the digestion chamber. We have also studied the binding of substrates to ClpAP, using the phage protein RepA, a 72-kDa dimer. RepA initially binds near the center of the distal ring of ClpA domains and may subsequently be visualized inside ClpP. The latter experiments employed a ClpP mutant which allows substrate translocation, but not its degradation. We have also performed more preliminary studies on the Lon enzyme of E. coli, which has the unusual feature that its protease and the ATPase are covalently linked as different domains of the same polypeptide chain, and on DFG1, a related ATPase from yeast. (2) Mechanotransduction by Myosin 1C from Acanthamoeba, a Single-headed Non-filamentous Myosin. From a biochemical and molecular biological perspective, AMIC is arguably the best characterized example of a single-headed myosin. In FY98, we started to study the structural basis of its interactions with actin filaments in different nucleotide states, hoping to elicit information that would also contribute to an improved understanding of muscle function. Our initial focus was on regulation of AMIC?s ATPase activity by heavy chain phosphorylation, taking advantage of specific point mutations in serine-329, the phosphorylable residue. The mutants are respectively, either constitutively active (Glu-329) or constitutively inactive (Ala-329). The mutant AMICs were found to be indistinguishable in their mode of binding to actin, implying that the phosphorylation-mediated regulatory event must affect some other aspect of the force-transducing cycle ? possibly molecular breathing of actin-bound AMIC. During FY99, this study was completed and the results were published. Our current objective is to use cryo-microscopy to examine the interactions of AMIC with phospholipid vesicles (its inferred activity in vivo involves vesicular transport). (3) Roles of the Umu Mutasome Complex in Recombination and Repair. The heterotrimeric UmuD?2C protein of E. coli plays a pivotal role in the SOS response of mutagenic DNA repair and may also inhibit homologous recombination. In both functions, UmuD?2C interacts with RecA nucleoprotein filaments. In this new project initiated in FY99, we studied this binding by electron microscopy, under both saturating and sub-saturating conditions. As estimated by a gel retardation assay, binding saturates at a stoichiometry of one complex per two RecA monomers. Cryo-microscopy under these conditions showed UmuD?2C to bind uniformly along the RecA filaments, such that the complexes are completely submerged in the deep helical groove. This mode of binding would impede access of other molecules to DNA in a RecA filament, thus explaining the ability of UmuD?2C to inhibit homologous recombination. On the other hand, SOS mutagenesis takes place in cells under conditions of sub-saturating binding. We examined the distribution of UmuD?2C complexes along RecA-ssDNA filaments by immuno-gold labelling with anti-UmuC antibodies, and found preferential binding at filament ends, accompanied by random binding to other sites at lower average occupancy. End-specific binding is consistent with genetic models whereby such binding appropriately positions UmuD?2C for the encounter with DNA polymerase III that leads to error-prone translesion DNA synthesis.(4) Structure and Assembly of Cornified Cell Envelopes (CEs) in Terminally Differentiated Keratinocytes. The CE is a covalently cross-linked layer of protein that lines the cytoplasmic surface of terminally differentiated keratinocytes in the epidermis and other squamous stratifying epithelia. CEs are thought to play a major role in conferring the physical resilience and impenetrability of these tissues. We have a long-term interest in the biogenesis of CEs, whose covalently cross-linked character has thwarted conventional biochemical approaches. In previous work, we applied a variety of electron microscopic approaches to CEs, both isolated and in situ, and combined these observations with compositional data acquired by mathematical modeling of their amino acid compositions. In this way, we developed a molecular model of the CE as a monolayer of molecules of the protein, loricrin, cross-linked both directly and via minor CE proteins. Thus we envisage the CE as a composite biomaterial with a matrix substance (loricrin) and cross-linkers (the minor proteins). By analogy with other composites it should be possible to modulate the mechanical properties of the CE by adjusting the ratio of loricrin to crosslinkers. Two main lines of enquiry were pursued in FY99. (a) Our colleagues at Baylor College of Medicine have created loricrin knockout mice. These animals are essentially normal ? a surprising finding in view of the important role attributed to loricrin in CE formation. While they lack loricrin, the knockout mice nevertheless have CEs. We have studied their structure and composition by the methods outlined above and find their CEs to be quasi-normal in every respect measured, expect for a somewhat altered amino-acid composition (hence, protein composition). We infer that there is a back-up system which supplies a hitherto unknown protein with loricrin-like properties that serves as a loricrin substitute in CE assembly in the knockout mice.(b) We have also studied epidermal samples from transgenic mice expressing a mutant form of loricrin that resembles, in its abnormal C-terminus, the form of the protein encountered in human patients with Vohwinkel syndrome and Progressive Symmetric Erythrokeratoderma. Electron microscopy reveals unusual deposits of the mutant loricrin in the cytoplasm and nucleus of granular layer cells and that deposits are not redispersed in corneocytes, as are L-granules containing normal loricrin. The presence of these deposits correlates with the phenotypes of generalized ichthyosis and thickening of footpad epidermis that are detected in these animals.